SYSTEMS AND METHODS FOR CONTOURING A SET OF MEDICAL IMAGES

• US 20190012567A1
• Filed: 10/16/2017
• Published: 01/10/2019
• Est. Priority Date: 05/03/2010
• Status: Active Grant

• Current Assignee: MIM Software Incorporated
• Original Assignee: MIM Software Incorporated
• Inventors: Piper, Jonathan William
• Granted Patent: US 10,599,948 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/174 involving the use of two or more images
• CPC Class Codes: G06V 10/755 Deformable models or variational models, e.g. snakes or active contours

Abstract

Systems and methods are provided for contouring a set of medical images. Deformation field data is generated between a source image and a target image of the set of medical images. The deformation field data relates structures in the source image to corresponding structures in the target image and is generated in accordance with a deformable registration algorithm. The deformation field data is utilized to generate target contour data associated with the target image from source contour data, associated with the source image, that identifies one or more objects within the source image.

Claims

1. A method of contouring a three-dimensional (3D) medical image, comprising:

receiving a source image slice, from a set of two-dimensional (2D) image slices constituting the 3D medical image, and source contour data identifying one or more objects within the source image slice;

receiving a selection of a target image slice, from the set of the 2D image slices;

generating deformation field data from the source image slice and the target image slice in accordance with a deformable registration algorithm, the deformation field data correlates locations in the source image slice to related locations in the target image slice; and

generating target contour data that identifies the one or more objects within the target image slice based on the source contour data and the deformation field data.

2. The method of claim 1, further comprising generating contour data for a subset of the set of 2D image slices constituting the 3D image by:

identifying a next target image slice in the set of 2D image slices to contour;

generating new deformation field data correlating locations in the target image slice to related locations in the next target image slice; and

applying the new deformation field data to the target contour data to generate new target contour data that identifies the one or more objects within the next target image slice.

3. The method of claim 2, further comprising utilizing the next target slice and new target contour data to successively generate contour data for another image slice of the subset.

4. The method of claim 2, further comprising associating contour data of image slices in the subset to generate a 3D contour for the 3D medical image.

5. The method of claim 1, further comprising generating contour data for a second target image slice, of the set of 2D image slices, by selecting the target image slice and target contour data as a new source image slice and new source contour data, respectively.

6. The method of claim 1, wherein the source image slice and the target image slice are adjacent image slices of the 3D medical image.

7. The method of claim 1, wherein the source image slice and the target image slice are non-adjacent image slices of the 3D medical image.

8. The method of claim 7, wherein generating the deformation field data comprises directly registering the source image slice to the target image slice with the deformation algorithm.

9. The method of claim 7, wherein generating the deformation field data comprises:

generating first intermediary deformation field data between the source image slice and an intermediary image slice, the intermediary image slice being a slice of the 3D medical image located between the source image slice and the target image slice;

generating second intermediary deformation field data between the intermediary image slice and the target image slice; and

concatenating the first intermediary deformation field data and the second intermediary deformation field data to generate the deformation field data between the source image slice and the target image slice.

10. The method of claim 1, wherein the source contour data is generated with manual contouring software.

11. The method of claim 1, further comprising:

receiving user input; and

modifying the target contour data in accordance with the user input.

12. The method of claim 1, further comprising:

identifying a predicted image slice from the set of 2D image slices predicted to be contoured;

generating predictive deformation field data from the target image slice to the predicted image slice;

receiving a selection of a next target image slice from the set of 2D image slices; and

utilizing the predictive deformation field data to generate next target contour data for the next target image slice when the predicted image slice and the next target image slice are identical.

13. The method of claim 1, wherein the deformable registration algorithm is an intensity-based deformation algorithm.

14. The method of claim 13, wherein the deformation field data identifies correspondences between pixels in the source image slice and pixels in the target image slice based on pixel intensity.

15. The method of claim 1, wherein the deformable registration algorithm is a free-form deformation algorithm.

16. The method of claim 1, wherein generating the target contour data comprises identifying, with the deformation field data, pixel positions in the target image slice corresponding to pixel positions in the source image slice of the one or more objects as identified by the source contour data.

17. The method of claim 1, wherein the source contour data and the target contour data comprise respective binary mask data respectively identifying whether pixels of the source image slice and the target image slice are inside or outside the one or more objects.

18. The method of claim 1, wherein the source contour data and the target contour data comprise respective byte mask data respectively identifying whether pixels of the source image slice and the target image slice are inside or outside the one or more objects.

19. A system for contouring a 3D medical image, comprising:

at least one processor; and

a non-transitory, computer-readable storage medium having stored thereon computer-executable instructions, the computer-executable instructions comprising;

an image deformation engine configured to;

receive a source image slice and a target image slice from a set of 2D image slices constituting the 3D medical image, and

generate deformation field data from the source image slice and the target image slice in accordance with deformable registration algorithm, the deformation field data correlates locations in the source image slice to related locations in the target image slice; and

a contour transformation engine configured to;

receive source contour data identifying one or more objects within the source image slice, and

generate target contour data that identifies the one or more objects within the target image slice based on the source contour data and the deformation field data.

20. The system of claim 19, wherein the image deformation engine and the contour transformation engine are further configured to contour additional image slices of the 3D medical image through utilization of the target image slice as a next source image slice and target contour data as next source contour data and selection of a next target image slice from the set of 2D image slices.